FORMATION OF CRYSTALS—GAUBERT. 275 
claims that the capillary action existing between the liquid and 
the crystal intervenes, an effect varying with the nature of the faces 
belonging to the diverse forms and with the nature of the liquid. 
Basing his belief on Gauss’s theory of capillarity, he concludes that 
such faces develop or require the minimum expenditure of capillary 
energy. The dominant forms must consequently be conditioned by 
those faces the constant capillarity of which is the least. The addi- 
tion of a foreign substance altering the different capillary constants 
may consequently induce modifications of form. 
It appears, indeed, that the capillary forces must act, but up to 
this time there is no fact known which proves that they intervene 
sufficiently to modify the forms, in spite of the experiments of 
M. Berent carried out in the laboratory of Sohncke; moreover, I shall 
describe later an observation showing they are without influence. 
le 
The crystals of one substance rarely form synchronously with those 
of another dissolved in the same mother liquid, and it is on this 
property that chemists base their action when they attempt to purify 
bodies by repeated crystallizations; but there are exceptions, as in 
the well-known coloration of hydrated nitrate of strontium by ex- 
tract of logwood, which was accomplished by Senarmont. Since 
then M. Lehmann and I have proved a few other cases of coloration 
of crystals by artificial organic dyes. 
By making use of the artificial coloration of crystals so as to indi- 
cate the presence of foreign matter which has crystallized with the 
colorless substance I have been enabled to show that the absorption 
caused modification in form. 
The absorption of foreign matter by crystals in process of formation 
is accomplished in two different ways: First, the coloring matter en- 
ters into the composition of the crystal, whatever may be its degree 
of dilution, and is shared between the crystal and the liquid; second, 
the coloring matter is taken up by the crystal only when the liquid 
becomes saturated. 
The two processes may go on simultaneously. The study of certain 
properties of colored crystals, particularly polychroism, and the law 
of division, shows that the coloring substance in the first case is found 
in the crystal in the same state as in the liquid; in the second, on the 
contrary, the coloring matter is in the crystalline state, and we have 
to do then with a regular grouping of the crystalline particles of the 
colorless substance with those of the coloring material added to the 
mother liquor. 
Lead nitrate is colored by methylene blue in the second manner; 
it appears in cubic crystals with the triglyphic strie of pyrite in- 
